Diabetic Retinopathy Possibly Results From Poor Blood Sugar Control Associated With MTHFR Gene Polymorphism in Type 2 Diabetic Patients: Response to Yoshioka et al.

We appreciate the comments of Dr. Yoshioka et al. (1). As described previously (2), we excluded the patients with >133 μmol/l serum creatinine level. In addition, the patients with >300 mg/dl urinary protein levels did not participate in our study. We considered that these exclusions must elucidate the effects of the methylenetetrahydrofolate reductase (MTHFR) gene polymorphism, not the effects of nephropathy, on the progression of diabetic retinopathy (DR) in type 2 diabetic patients. We agree with their comment that we analyzed the correlation with a smaller number of subjects. However, we cannot help referring to the difference in the backgrounds of the subjects between the two studies. In our study, the subjects had a mean age of 59.4 years, a mean diabetes duration of 10.8 years, a mean HbA_1c of 8.1%, and a mean serum creatinine of 0.76 mg/dl. The noteworthy difference between the two studies is the mean HbA_1c level (8.1 vs. 7.3%). The discrepancy may be attributable to this difference.

To support this hypothesis, the subjects with >9.8% HbA_1c level were excluded from the previous analysis (2) to get the mean HbA_1c level down to 7.3%, and then the data were analyzed again. As a result, there was no significant difference in the relationship between the MTHFR gene polymorphism and DR (n = 124, χ² test, P = 0.08). Fong et al. (3) described that epidemiological analysis of the U.K. Prospective Diabetes Study data showed a continuous relationship between the risk of microvascular complications and glycemia, such that for every percentage point decrease in HbA_1c (e.g., from 8 to 7%), there was a 35% reduction in the risk of microvascular complications. Based on this description, improved control of blood glucose may mask the retinopathic background derived from the MTHFR gene polymorphism. Thus, in this letter, we propose that the MTHFR gene polymorphism contributes to the progression of DR synergistically with impaired blood glucose control. In other words, blood glucose control could override the effects of the MTHFR gene polymorphism in type 2 diabetic patients.

Prospective cohort studies are required to understand the influence of the MTHFR gene polymorphism on the progression of DR. We thank Yoshioka et al. again for their comment, which has illuminated that blood glucose control may be associated with the effect of the MTHFR gene polymorphism on DR.